AI Paternalism System
(Redirected from Benevolent AI Control System)
Jump to navigation
Jump to search
An AI Paternalism System is an AI control system that makes decisions for human users based on presumed superior knowledge.
- AKA: Paternalistic AI System, AI Guardian System, Benevolent AI Control System.
- Context:
- It can typically override user preferences with optimization algorithms.
- It can typically justify interventions through utility maximization.
- It can often create dependency relationships with passive users.
- It can often reduce deliberative capacity through decision offloading.
- It can range from being a Soft AI Paternalism System to being a Hard AI Paternalism System, depending on its override strength.
- It can range from being a Narrow AI Paternalism System to being a Comprehensive AI Paternalism System, depending on its domain scope.
- It can range from being a Transparent AI Paternalism System to being a Opaque AI Paternalism System, depending on its decision visibility.
- It can range from being a Reversible AI Paternalism System to being a Irreversible AI Paternalism System, depending on its user control.
- ...
- Examples:
- Healthcare AI Paternalism Systems, such as:
- Financial AI Paternalism Systems, such as:
- ...
- Counter-Examples:
- AI Decision Support System, which provides recommendations without enforcement.
- Human Autonomy Preservation System, which maintains user agency.
- Collaborative AI System, which shares decision authority with humans.
- See: Paternalism, AI Control System, Soft Despotism, Human Autonomy Preservation Task, AI Dependency Risk, Omniscient Autocomplete Thought Experiment, Nudge Architecture, Behavioral Economics, Choice Architecture.